Physics-guided transfer learning for Bayesian optimization of chemical port-Hamiltonian systems
Authors
Negareh Mahboubi, Junyao Xie, Biao Huang
Abstract
Bayesian optimization (BO) has emerged as a powerful black-box optimization approach for complex systems, making sequential decisions through Gaussian process (GP) models to explore complex search spaces. However, conventional BO faces certain challenges when applies to optimizations of chemical systems, particularly with limited measurement data and physical constraints. This paper proposes an adaptive framework combining transfer learning with physics-informed GP to enhance BO performance for chemical process optimization. By incorporating physics-based priors through Gaussian Process Port-Hamiltonian Systems (GP-PHS) in the point-by-point transfer learning methodology, the proposed approach dynamically leverages knowledge from related source domains while satisfying physical constrains. The framework’s effectiveness is demonstrated across three chemical systems including a water tank, an electrochemical cell, and an isothermal continuous stirred tank reactor (CSTR). Results show improvements in both optimization accuracy and convergence speed compared to traditional BO methods. This proposed approach bridges the gap between data-driven optimization and physical principles, offering a robust solution for complex chemical system optimization under data scarcity.
Keywords
Bayesian optimization; Transfer learning; Gaussian process; Port-Hamiltonian systems; Physics-enhanced machine learning
Citation
- Journal: Computers & Chemical Engineering
- Year: 2025
- Volume: 203
- Issue:
- Pages: 109331
- Publisher: Elsevier BV
- DOI: 10.1016/j.compchemeng.2025.109331
BibTeX
@article{Mahboubi_2025,
title={{Physics-guided transfer learning for Bayesian optimization of chemical port-Hamiltonian systems}},
volume={203},
ISSN={0098-1354},
DOI={10.1016/j.compchemeng.2025.109331},
journal={Computers & Chemical Engineering},
publisher={Elsevier BV},
author={Mahboubi, Negareh and Xie, Junyao and Huang, Biao},
year={2025},
pages={109331}
}
References
- Agrell, (2019)
- Bai, (2023)
- Bard, (2022)
- Beckers, Data-driven Bayesian control of port-Hamiltonian systems. (2023)
- Beckers, Gaussian process port-Hamiltonian systems: Bayesian learning with physics prior. (2022)
- Brochu, (2010)
- Cardoso-Ribeiro, Port-Hamiltonian modeling, discretization and feedback control of a circular water tank. (2019)
- Cardoso-Ribeiro, F. L., Matignon, D. & Lefèvre, L. Dissipative Shallow Water Equations: a port-Hamiltonian formulation. IFAC-PapersOnLine 54, 167–172 (2021) – 10.1016/j.ifacol.2021.11.073
- Cheon, M., Byun, H. & Lee, J. H. Non-myopic Bayesian optimization using model-free reinforcement learning and its application to optimization in electrochemistry. Computers & Chemical Engineering 184, 108624 (2024) – 10.1016/j.compchemeng.2024.108624
- Deshwal, A., Simon, C. M. & Doppa, J. R. Bayesian optimization of nanoporous materials. Mol. Syst. Des. Eng. 6, 1066–1086 (2021) – 10.1039/d1me00093d
- Di Fiore, F. & Mainini, L. Physics-aware multifidelity Bayesian optimization: A generalized formulation. Computers & Structures 296, 107302 (2024) – 10.1016/j.compstruc.2024.107302
- Dirksz, Adaptive tracking control of fully actuated port-Hamiltonian mechanical systems. (2010)
- Evangelisti, Physically consistent learning of conservative Lagrangian systems with Gaussian processes. (2022)
- Fan, (2023)
- Feurer, Hyperparameter optimization. Autom. Mach. Learn.: Methods Syst. Chall. (2019)
- Feurer, (2018)
- Feurer, (2018)
- Frazier, (2018)
- Frazier, Bayesian optimization for materials design. (2015)
- González, L. D. & Zavala, V. M. New paradigms for exploiting parallel experiments in Bayesian optimization. Computers & Chemical Engineering 170, 108110 (2023) – 10.1016/j.compchemeng.2022.108110
- Hjelmfelt, A., Weinberger, E. D. & Ross, J. Chemical implementation of neural networks and Turing machines. Proc. Natl. Acad. Sci. U.S.A. 88, 10983–10987 (1991) – 10.1073/pnas.88.24.10983
- Hoang, H., Couenne, F., Jallut, C. & Le Gorrec, Y. The port Hamiltonian approach to modeling and control of Continuous Stirred Tank Reactors. Journal of Process Control 21, 1449–1458 (2011) – 10.1016/j.jprocont.2011.06.014
- Hou, Z.-S. & Wang, Z. From model-based control to data-driven control: Survey, classification and perspective. Information Sciences 235, 3–35 (2013) – 10.1016/j.ins.2012.07.014
- Ikonen, (2001)
- Jidling, (2017)
- Jones, D. R., Schonlau, M. & Welch, W. J. Efficient Global Optimization of Expensive Black-Box Functions. Journal of Global Optimization 13, 455–492 (1998) – 10.1023/a:1008306431147
- Theckel Joy, T., Rana, S., Gupta, S. & Venkatesh, S. A flexible transfer learning framework for Bayesian optimization with convergence guarantee. Expert Systems with Applications 115, 656–672 (2019) – 10.1016/j.eswa.2018.08.023
- Kandasamy, (2019)
- Karniadakis, G. E. et al. Physics-informed machine learning. Nat Rev Phys 3, 422–440 (2021) – 10.1038/s42254-021-00314-5
- Lamparth, (2022)
- Li, Y. et al. OpenBox: A Generalized Black-box Optimization Service. Proceedings of the 27th ACM SIGKDD Conference on Knowledge Discovery & Data Mining 3209–3219 (2021) doi:10.1145/3447548.3467061 – 10.1145/3447548.3467061
- Li, P., Tan, K. & Beckers, T. PyGpPHs: A Python Package for Bayesian Modeling of Port-Hamiltonian Systems. IFAC-PapersOnLine 58, 54–59 (2024) – 10.1016/j.ifacol.2024.08.256
- Ma, (2019)
- Mahboubi, Point-by-point transfer learning for Bayesian optimization: An accelerated search strategy. Comput. Chem. Eng. (2024)
- Maraval, (2023)
- McLachlan, On the number of components in a Gaussian mixture model. Wiley Interdiscip. Rev.: Data Min. Knowl. Discov. (2014)
- Min, A. T. W., Ong, Y.-S., Gupta, A. & Goh, C.-K. Multiproblem Surrogates: Transfer Evolutionary Multiobjective Optimization of Computationally Expensive Problems. IEEE Trans. Evol. Computat. 23, 15–28 (2019) – 10.1109/tevc.2017.2783441
- Močkus, J. On Bayesian Methods for Seeking the Extremum. Optimization Techniques IFIP Technical Conference 400–404 (1975) doi:10.1007/978-3-662-38527-2_55 – 10.1007/978-3-662-38527-2_55
- Mockus, The application of Bayesian methods for seeking the extremum. (2014)
- Moosavi, (2022)
- Murphy, (2012)
- Nageshrao, S. P., Lopes, G. A. D., Jeltsema, D. & Babuska, R. Port-Hamiltonian Systems in Adaptive and Learning Control: A Survey. IEEE Trans. Automat. Contr. 61, 1223–1238 (2016) – 10.1109/tac.2015.2458491
- Perrone, Learning search spaces for Bayesian optimization: Another view of hyperparameter transfer learning. Adv. Neural Inf. Process. Syst. (2019)
- Rajabi-Kochi, M., Mahboubi, N., Gill, A. P. S. & Moosavi, S. M. Adaptive representation of molecules and materials in Bayesian optimization. Chem. Sci. 16, 5464–5474 (2025) – 10.1039/d5sc00200a
- Ramachandran, Information-theoretic transfer learning framework for Bayesian optimisation. (2019)
- Rasmussen, (2005)
- Sbarbaro, D. On the Port-Hamiltonian Models of some Electrochemical Processes. IFAC-PapersOnLine 51, 38–43 (2018) – 10.1016/j.ifacol.2018.06.010
- Schilling, Scalable hyperparameter optimization with products of Gaussian process experts. (2016)
- Shahriari, B., Swersky, K., Wang, Z., Adams, R. P. & de Freitas, N. Taking the Human Out of the Loop: A Review of Bayesian Optimization. Proc. IEEE 104, 148–175 (2016) – 10.1109/jproc.2015.2494218
- Shields, B. J. et al. Bayesian reaction optimization as a tool for chemical synthesis. Nature 590, 89–96 (2021) – 10.1038/s41586-021-03213-y
- Snoek, (2012)
- Sussex, (2022)
- van der Schaft, A. & Jeltsema, D. Port-Hamiltonian Systems Theory: An Introductory Overview. FnT in Systems and Control 1, 173–378 (2014) – 10.1561/2600000002
- Wang, Y., Chen, T.-Y. & Vlachos, D. G. NEXTorch: A Design and Bayesian Optimization Toolkit for Chemical Sciences and Engineering. J. Chem. Inf. Model. 61, 5312–5319 (2021) – 10.1021/acs.jcim.1c00637
- Wang, Y., Feng, G. & Cheng, D. Simultaneous stabilization of a set of nonlinear port-controlled Hamiltonian systems. Automatica 43, 403–415 (2007) – 10.1016/j.automatica.2006.09.008
- Wang, T. et al. Edge Computing and Sensor-Cloud: Overview, Solutions, and Directions. ACM Comput. Surv. 55, 1–37 (2023) – 10.1145/3582270
- Williams, (2006)
- Williams, (2006)
- Wistuba, (2021)
- Wistuba, M., Schilling, N. & Schmidt-Thieme, L. Scalable Gaussian process-based transfer surrogates for hyperparameter optimization. Mach Learn 107, 43–78 (2017) – 10.1007/s10994-017-5684-y
- Xie, J., Huang, B. & Dubljevic, S. Transfer Learning for Dynamic Feature Extraction Using Variational Bayesian Inference. IEEE Trans. Knowl. Data Eng. 34, 5524–5535 (2022) – 10.1109/tkde.2021.3054671
- Yang, L., Zhang, X., Lu, J., Tian, Z. & Du, W. A local space transfer learning-based parallel Bayesian optimization with its application. Chinese Journal of Chemical Engineering 74, 227–237 (2024) – 10.1016/j.cjche.2024.04.029